Bridging the Temporal Gap in Multimodal LLMs: Deeply Stacking Temporal Tokens for Audio-Visual Speech Recognition

摘要

Audio-Visual Speech Recognition enhances speech recognition robustness in noisy conditions by leveraging visual cues. However, current Multimodal LLMs suffer from a fundamental temporal gap. This gap is characterized by limited fine-grained temporal modeling in vision encoders and progressive temporal semantic degradation throughout the deep layers of LLM decoders. To bridge this gap, we propose a novel framework that deeply stacks temporal tokens across both the encoding and decoding stages. Specifically, we enhance the vision encoder with a temporal-aware attention module and temporal rotary positional embeddings to precisely capture the sequential evolution and dynamics of lip movements. Furthermore, we stack hierarchical temporal tokens that incorporate temporally enriched features into multiple layers of the LLM decoder in a bottom-up manner. Extensive experiments on the LRS2 and LRS3 benchmarks demonstrate that our approach achieves high efficiency and firm performance, outperforming existing supervised, self-supervised, and LLM-based methods by 6.1% on LRS2 and 7.8% on LRS3.